Department of Psychology, University of Waterloo, 200 University Avenue West, Waterloo, Ontario, N2L 3G1, Canada.
Body Image. 2022 Dec;43:41-53. doi: 10.1016/j.bodyim.2022.08.005. Epub 2022 Aug 24.
Findings have been mixed as to whether individual differences in within-person variability in body image predict maladaptive body image and eating behaviors. The current study aimed to resolve this ambiguity by addressing limitations of past research. First, we measured within-person variability in body image across the context-sensitive domain of relationships. Second, we incorporated the latest statistical methods to increase the robustness of the results. Online, 189 female-identified undergraduates completed seven baseline measures of trait body image. At least three days later, in-lab, participants were guided to generate a list of the most important people in their lives (i.e., friends, family members, close others) using egocentric network methods. Participants then completed a set of three relationship-specific measures in which they reported on their typical body image with 10 people from their list, one by one. Multiverse analysis tested the hypothesis that, across combinations of measures, within-person variability in relational body image would positively predict indicators of maladaptive body image. In 84 regression analyses, permutation testing supported our overall hypothesis (p = .006); however, results varied across different model specifications. Results provide further evidence for the predictive power of within-person variability in body image and yield valuable methodological and statistical recommendations.
关于个体在身体意象上的个体内变异性是否能预测适应不良的身体意象和饮食行为,研究结果喜忧参半。本研究旨在通过解决过去研究的局限性来解决这一模糊性。首先,我们在关系敏感的领域内测量了身体意象的个体内变异性。其次,我们采用了最新的统计方法来提高结果的稳健性。189 名女性身份的本科生在线完成了七项身体意象特质的基线测量。至少三天后,在实验室里,参与者被引导使用自我中心网络方法生成一份他们生活中最重要的人的名单(即朋友、家人、亲密的人)。然后,参与者完成了一组三个关系特定的测量,他们逐个用名单上的 10 个人报告他们的典型身体形象。多元宇宙分析检验了这样一个假设,即在各种测量组合中,关系身体意象的个体内变异性将积极预测适应不良身体意象的指标。在 84 项回归分析中,置换检验支持了我们的总体假设(p=0.006);然而,结果在不同的模型规范中有所不同。研究结果进一步证明了身体意象个体内变异性的预测能力,并提供了有价值的方法学和统计学建议。